OpenAI Retail Asia-Pacific & Oceania

avatarin

Customer Service Automation · Sales Automation

avatarin needed to extend expert retail support beyond store hours and turn product discovery into a natural multilingual conversation instead of a brittle keyword chatbot.

24/7 multilingual support by voice and text
30,000 shoppers engaged during a two-week public experience
92% of post-use survey responses were positive

Reported impact

Business outcome

avatarin proved that AI can extend retail expertise beyond store hours, delivering strong user satisfaction, large public engagement, and a foundation for one brand-consistent intelligence that can follow customers across channels.

  • Customer Experience
  • Revenue Growth
  • Scalability

Solution

avatarin used GPT-Realtime to build the Kurashi-Marugoto AI Agent for Yamada Denki, creating a 24/7 shopping assistant that can speak, listen, and reason over product information in one loop. The system combines low-latency voice interaction with a retrieval-augmented generation layer so shoppers can ask open-ended questions about needs, space constraints, or budget and still receive grounded product guidance. avatarin also translated Yamada Denki's category-specific sales expertise into prompts and conversation flows, allowing the agent to ask follow-up questions rather than waiting for exact keywords. Guardrails keep the interaction centered on shopping outcomes even when customers change direction, and OpenAI worked with the team on prompt structure, cost optimization, and implementation practices for always-on service. The result is a branded retail interface that preserves context across voice and text while capturing the intent signals conventional ecommerce channels often miss.

Data flow

A shopper speaks or types into the Kurashi-Marugoto AI Agent, which uses GPT-Realtime plus a retrieval-augmented generation layer to combine intent, product data, and Yamada Denki sales patterns into guided recommendations. Voice surveys at the end of conversations feed back customer reactions and intent signals for future improvement.

Solution architecture

4 components · 3 layers
  1. Compute
    • GPT-Realtime Provides low-latency multimodal conversation for voice, text, and image understanding.
    • OpenAI API Supports earlier speech recognition, inquiry analysis, and implementation patterns used in production.
  2. Serving
    • Kurashi-Marugoto AI Agent Acts as the shopper-facing retail assistant for product discovery and purchase guidance.
  3. Orchestration
    • retrieval-augmented generation system Grounds answers in product data while preserving responsive conversation.

Architecture clues

  • Conversation flows encode Yamada Denki sales expertise by product category
  • GPT-Realtime unifies speech, text, and image handling in one conversational loop
  • RAG grounds responses in product information
  • Voice surveys are captured at the end of each interaction

Evidence from the source

"92% of survey responses were positive"
"A retrieval-augmented generation system grounds the agent’s responses in product information"
"Built on OpenAI’s GPT‑Realtime, the Kurashi-Marugoto AI Agent supports natural voice conversations"
"approximately 30,000 people used the agent"
Transparency Evidence & methodology 100/100 reference value · 87% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value100/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth25/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact25/25
    2 metrics describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence87%
  • Source authority20/25
    The main evidence is an official vendor-published customer story.
  • Claim traceability21/25
    10 of 12 key fields link to a named source.
  • Direct evidence20/20
    4 direct source quotes retained.
  • Date provenance8/10
    The publication date comes from publisher-controlled metadata.
  • Source transparency8/10
    1 source reference document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. OpenAI

    avatarin customer story

    First-party
    https://openai.com/index/avatarin/
    Source type
    Vendor-published customer story
    Published
    Jul 30, 2026
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date

Date evidence

Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.